Energy efficiency and ventilation in Swedish industries

(1)

Linköping Studies in Science and Technology Dissertation No. 1223

Energy efficiency and ventilation in

Swedish industries

barriers, simulation and control strategy

Patrik Rohdin

Division of Energy Systems

Department of Management and Engineering

Linköping University, Linköping, Sweden

(2)

ISSN: 0345-7524

(3)

This thesis is based on work conducted within the interdisciplinary graduate school Energy Systems. The national Energy Systems Programme aims at creating competence in solving complex energy problems by combining technical and social sciences. The research programme analyses processes for the conversion, transmission and utilisation of energy, combined together in order to fulfil specific needs.

The research groups that participate in the Energy Systems Programme are the Department of Engineering Sciences at Uppsala University, the Division of Energy Systems at Linköping Institute of Technology, the Department of Technology and Social Change at Linköping University, the Division of Heat and Power Technology at Chalmers University of Technology in Göteborg as well as the Division of Energy Processes at the Royal Institute of Technology in Stockholm.

(4)

(5)

Abstract

The energy issue is presently in focus worldwide. This is not only due to increasing environmental concern regarding energy related emissions, but also due to the trend of increasing energy prices. Energy usage in the industrial sector in Sweden today represents about one third of the national energy use. A substantial part of that is related to support processes such as heating, ventilation and cooling systems. These systems are important as they are related both to energy cost and indoor climate management as well as to the health of the occupants.

The purpose of this thesis is to reach a more comprehensive view on industrial energy efficiency and indoor environment issues related to industrial ventilation. This has been studied in three themes where the first part addresses barriers to energy efficiency in Swedish industries, the second theme discuss simulation as decision support, and the third studies the variable air volume system in industrial facilities.

In the first theme three different studies were made: the first studies non-energy intensive companies in Oskarshamn in Sweden, the second studies the energy intensive foundry industry and the third study was part of an evaluation of a large energy efficiency program called Project Highland. These studies had several findings in common, such as the importance of a strategic view on the energy issue and the presence of a person with real ambition with power over investment decisions related to energy issues at the company. The studies also show that several information related barriers are important for decision makers at the studied companies. This shows that information related barriers are one reason in why energy efficient equipment is not implemented.

In the second theme the use of simulation in the form of Computational Fluid Dynamics (CFD) and Building Energy Simulation (BES) are used as decision support for industrial ventilation related studies at two different industries, one foundry is investigated and one dairy. BES has mainly been used to simulate energy and power related parameters while CFD was used to give a detailed description of the indoor and product environment. Together these methods can be used to better evaluate the energy, indoor and product environment and

(6)

thus enable the implementation of more efficient heating, ventilation and air-conditioning systems.

In the third theme the use of Variable Air Volume (VAV) systems was evaluated, and was found to be an efficient way to reduce energy use at the studied sites. At the studied foundry the VAV system is predicted to reduce space heating and electricity use by fans by about 30%, and in the dairy case by about 60% for space heating and 20% for electricity.

(7)

Sammanfattning

Energifrågan är idag högaktuell, dels beroende på kopplingen till hotet om global uppvärmning, dels beroende på stigande energipriser. Den industriella energianvändningen i Sverige står för ca en tredjedel av den totala energianvändningen och en betydande del av denna användning är relaterad till stödprocesser som till exempel luftkonditionering, värmning, kylning och ventilation. Dessa processer är viktiga då de är kopplade till såväl energianvändning som hälsa och inomhusmiljö.

Syftet med denna avhandling är att ge en mångsidig beskrivning av industriell energieffektivisering och industriventilation. Detta har studerats i tre teman där det första temat tar upp frågan om hinder mot energieffektivisering, det andra temat simulering som beslutsstöd och det tredje temat variabelflödessystem för industriell ventilation.

I det första temat har tre studier genomförts, en studie av icke energiintensiva företag i Oskarshamn, en studie av den energiintensiva gjuteriindustrin i Sverige och en studie i samband med en utvärdering av Projekt högland. Dessa studier har flera gemensamma nämnare som exempelvis behovet av en långsiktig energistrategi samt behovet av en eldsjäl med makt över investeringsbeslut. Dessa studier visar också på flera informationsrelaterade hinder som är viktiga för beslutsfattarna på företagen. Det andra temat undersöker hur simulering, i form av Computational Fluid Dynamics (CFD) och byggnadssimulering (BES) kan användas som beslutsstöd. Två detaljerade studier har genomförts inom detta tema, ett lättmetallgjuteri och ett mejeri har studerats. Byggnadssimulering har främst använts för att studera effekt och energirelaterade parametrar medan CFD har använts för att ge en detaljerad beskrivning av termisk- och produktmiljö. Tillsammans ger dessa båda metoder bättre möjligheter att utvärdera energi, produkt och inomhusmiljörelaterade aspekter, vilket ger bättre möjligheter att dimensionera effektivare system. I det tredje temat har potentialen för att använda variabelflödessystem undersökts, och för de studerade anläggningarna finns en stor potential för tekniken. För den studerade gjuterilokalen visar simuleringarna på en reducering av värmning och fläktel med ca 30% och för den studerade mejerianläggningen en reducerad värmeanvändning med ca 60% och elanvändning med ca 20%.

(8)

Acknowledgements

irst, I would like to thank my supervisor, Professor Bahram Moshfegh, both for introducing me to the world of research and for all the help, encouragement and valuable input on this thesis. I would also like express my gratitude to him for encouraging me to do so many different projects and studies, as this has made my time as a PhD student both interesting and rewarding. I would also like to thank my assistant supervisor, Professor Hazim Awbi, at Reading University, U.K. for the support and comments on drafts of this thesis and valuable comments in the early parts of my PhD project. Furthermore, I would like to thank my assistant supervisor at the Division, Dr. Magnus Karlsson, for valuable comments on drafts as well as fruitful collaboration on several projects, and Dr. Jenny Palm for all the help with the interview guide and for reading drafts of this thesis.

A special thanks also to the co-authors with whom I have had the privilege to work. Ph.D. student Patrik Thollander for all the work we did together during the first years of our PhD studies, not to mention for showing me that you can actually fish even though Linköping is far from the ocean. Dr. Fredrik Karlsson for all the work we did together on various projects as well as all the sometimes very long discussions. Our collaboration has meant a lot to me. Ph.D. student Maria Danestig, Dr. Marie-Louise Persson and Mr Petter Solding for all the work we did together and Tekn. Lic. Ulf Larsson and Dr. Mathias Cehlin at the University of Gävle, for the CFD discussions. The representatives at the companies I have worked with are also gratefully acknowledged. Mr Sten-Erik Lindhe, Mr Lars-Erik Stöllman, Mr Kjell Lunden-Pettersson, Mr Hans Nycander and Mr Hans-Olov Appelgren at Arla Foods AB and Mr Gunnar Göransson, Mr Anders Gustafsson and Mr Ola Ring at Husqvarna AB to name a few. Thank you.

I would also like to thank all my colleagues at the Division of Energy Systems in Linköping as well as all other colleagues within the National Energy Systems Programme. Without you my time as a PhD student would have been much less rewarding, not to mention fun.

Finally, I would like to thank my friends and family, and especially my fiancée Johanna, for the things in life that matter the most.

(9)

Appended papers

Theme 1

1. Rohdin, P. Thollander, P. (2006) Barriers to and Driving Forces for Energy Efficiency in the Non-Energy Intensive Manufacturing Industry in Sweden. Energy 31 1500-1508.

2. Rohdin, P. Thollander, P. Solding, P. (2007) Barriers to and Drivers for Energy Efficiency in the Swedish Foundry Industry. Energy Policy 35 672-677.

3. Thollander, P. Danestig, M. Rohdin, P. (2007) Energy Policies for Increased Industrial Energy Efficiency - Evaluation of a Local Energy Programme for Manufacturing SMEs, Energy Policy 35 5774-5783.

Theme 2

4. Rohdin, P. Moshfegh, B. (2007) Numerical Predictions of Indoor Climate in Large Industrial Premises - A Comparison between Different

k-ε Models Supported by Field Measurements. Building and Environment

42 3872-3888.

5. Rohdin, P. Moshfegh, B. (2006) Numerical Predictions of Indoor Climate in a Light Alloy Casting Facility. Proc. of the 13th International

Heat Transfer Conference, IHTC-13, August 13-18, Sydney, Australia

6. Karlsson, F. Rohdin, P. Persson, M-L. (2007) Measured and Predicted Energy Demand of a Low Energy Building - Important Aspects When Using Building Energy Simulation. Building Services Engineering Research and

Technology 28 1-13.

Theme 3

7. Rohdin, P. Moshfegh, B. (2007) CFD and BES as Decision Support When Implementing a Variable Air Volume System in a Foundry. In Proc. of Clima 2007 WellBeing Indoor Congress, 10-14 June, Helsinki, Finland.(The paper was awarded with a Best Poster Honorary Mention)

8. Rohdin, P. Moshfegh, B. (2007) Variable Air Volume-Flow Systems – A Possible Way to Reduce Energy Use in the Swedish Dairy Industry.

(10)

Nomenclature

C Concentration (ppm)

p

C Specific heat, (J/kg K)

fcl Clothing surface factor (-)

gi Component of the gravitational vector in the ith direction (m/s2)

hc Convective heat transfer coefficient (W/m2K)

Icl Clothing insulation (m2K/W)

k Turbulent kinetic energy, (m2_/s2₎

l Length scale (m)

M Metabolic rate (W)

Pfans Power used by fans (kW)

Pa Water vapor partial pressure (N/m2) Pr Prandtl number (-)

p Pressure, (N/m2₎

q Airflow (m3_/s)

Sij Magnitude of rate of strain (1/s) TI Turbulence intensity, (-)

t Temperature (°C)

ta Air temperature (°C)

tcl Clothing surface temperature (°C)

r

t Mean radiant temperature (°C)

U Mean velocity, (m/s)

u~ Instantaneous velocity, (m/s)

u Velocity fluctuation, (m/s)

V Volume (m3₎

vair Relative air velocity (m/s)

v& Ventilation rate (ls-1olf-1)

W Effective mechanical power (W)

y Distance to wall (m)

y+ _{Dimensionless wall distance (-)}

(11)

Greek symbols

β Volumetric thermal expansion coefficient (1/K) δ Kronecker delta, (-)

ε Dissipation of turbulent kinetic energy, (m2_/s3₎

λ Thermal conductivity, (W/mK) μ Dynamic viscosity, (kg/m s) μt Eddy viscosity (kg/m s) ν Kinematic viscosity, (m2_/s) νt Eddy viscosity, (m2/s) ρ Density, (kg/m3₎ θ~ Instantaneous temperature (°C), θ~=Θ+θ Θ Average temperature, (°C, K) 0 Θ Reference temperature (°C, K) θ Temperature fluctuation (°C, K) t

σ Turbulent Prandtl number (-)

ε

σ

σk, Constants

τi Mean age of air (s)

τn Nominal time constant (s)

τij Stress components (N/m2) Subscript

oz occupied zone i inlet

(12)

(13)

1 Introduction

This introduction outlines the background of this thesis followed by the purpose, aim and scope. The co-author statement as well as a short description of the appended papers are also included in this chapter.

he threat of global warming as a result of the use of fossil fuels and other energy-related emissions are forcing politicians on all levels are to try to influence energy use, and as a result different policy instruments have been implemented such as CO2 emissions trading. These and other future policy instruments will in all probability result in higher energy prices to decrease the use of fossil fuels, and thus further increase the demand for industrial energy efficiency. From a Swedish perspective the increased globalization and opening up of domestic markets within the European Union will also make the implementation of cost-efficient energy efficiency measures within industry even more necessary, due to among other things the historically low prices of electricity. This highlights the need for an increase in energy efficiency investments for Swedish industrial companies.

Energy usage in the industrial sector in Sweden today represents a large part of the national energy use, but is still considered by many Swedish companies to be a non-strategic factor. The structure of industrial energy use is complex, due to a high degree of interdependency among processes, technology development affecting e.g. energy efficiency and energy conservation, and various dynamics coming into play through production schedules, energy prices, raw materials, labor force and other management issues. From a corporate perspective this introduces multiple problems when working with energy-efficiency issues.

T

(16)

Support processes in industrial energy systems, such as ventilation, heating and cooling systems, are important issues in industrial premises as they are related both to energy cost and indoor climate management as well as to the health of the occupants. Heating, cooling and ventilation of industrial premises account for a large part of the total industrial energy usage in manufacturing. Poor indoor environment conditions in industries also cost large amounts of money in health care, administration, and lost productivity. This underscores the importance of well functioning Heating, Ventilation and Air-Conditioning (HVAC) systems.

1.1 Purpose

The aim of this thesis is to reach a more comprehensive view on energy performance and indoor environmental issues related to industrial ventilation. The structure of the thesis is summarized in five research questions:

1. Is there a potential for cost-efficient energy efficiency measures in Swedish industry?

2. What barriers inhibit the implementation of cost-efficient energy efficiency measures?

3. What barriers are specific to Heating, Ventilation and Air-Conditioning (HVAC) in industrial ventilation?

4. How can Building Energy Simulation (BES) and Computational Fluid Dynamics (CFD) be used to reduce these barriers to energy efficiency for HVAC in industrial premises?

5. Is there potential for using Variable Air Volume (VAV) systems in industrial premises?

There are three main themes within this thesis. The first theme focuses on barriers and drivers affecting the adoption of energy efficiency measures and is connected to the first three research questions. The second theme picks up issues derived from the first theme and focuses on Building Energy Simulation and Computational Fluid Dynamics as decision support. The third theme focuses on the effects of Variable Air Volume (VAV) control in industrial premises.

(17)

1.2 Scope of the thesis

This thesis primarily treats energy efficiency issues in industries with focus on decision support for industrial ventilation applications. All the studies are made mainly from a company perspective and the barriers to energy efficiency presented reflect the respondents’ at the studied companies view of the importance of the different barriers. The same perspective has been used in detailed studies at two large industrial companies where the energy efficiency potential reflects possible reductions at the facility or site. No studies of how these measures would affect the surrounding energy system have been made. This also includes the fact that no explicit studies of environmental performance have been made. Furthermore, when studying the indoor environment, secondary parameters such as sound, lighting, etc. have not been studied. The focus has been on air quality and thermal comfort and to some degree the product environment. The studies are made exclusively in a Swedish context.

1.3 Outline of the thesis

This thesis includes 11 chapters. Chapter 1 contains background, purpose, scope and brief description of the appended papers. Chapter 2 summarizes some aspects of Swedish industrial energy use and also shows some recent studies of the energy efficiency potential within the Swedish industrial sector. In Chapter 3 some industrial management aspects related to decision-making in industrial firms are also discussed and the concept of barriers to energy efficiency is presented. Chapter 4 describes some requirements and overall measures of performance used for industrial ventilation and HVAC in industrial premises, such as indoor air quality, thermal comfort, product environment and energy use aspects. Chapter 5 treats industrial ventilation and its aspects such as control systems, air supply issues, local ventilation and the measures of performance used. Chapter 6 treats the methods used in the appended papers including measurements, Computational Fluid Dynamics (CFD), Building Energy Simulation (BES) and qualitative case studies, including interviews and questionnaires. In Chapter 7, results from the studies of barriers and drivers to energy efficiency are presented. This chapter includes a presentation of the findings in the first three appended papers as well as a discussion of these barriers in relation to industrial ventilation. Chapter 8

(18)

contains a description of two detailed studies at one foundry and two dairy processing plants. The production processes, energy use structure as well as CFD and BES are presented. Chapter 9 sums up and discusses the main ideas and Chapter 10 concludes the thesis. In Chapter 11 further work is presented.

1.4 Co-author statement

The first two papers were planned, executed and written in co-operation with PhD student Patrik Thollander. The third paper was written in collaboration with Patrik Thollander and PhD student Maria Danestig, where the author mainly contributed with issues related to the barrier study as well as writing parts of the paper. Paper four was written mainly by the author, but professor Bahram Moshfegh wrote parts of the section about computational fluid dynamics in addition to making valuable comments on drafts of the paper. Paper five was written entirely by the author while Professor Bahram Moshfegh contributed with valuable input and comments on drafts. Paper six was written in collaboration with Dr. Fredrik Karlsson at Linköping University and Dr. Mari-Louise Persson at Ångström, Uppsala University. The author made simulations and exclusively wrote the part about one of the three software programs and was also responsible for the sensitivity analysis. The rest of the paper was co-written with the other authors. Papers seven and eight were written by the author of this thesis but Professor Bahram Moshfegh contributed with many important comments on drafts and ideas as well as made valuable contributions when the measuring strategy for the papers was formulated.

(19)

1.5 Appended papers in brief

Rohdin, P. Thollander, P. (2006) Barriers to and Driving Forces for Energy Efficiency in the Non-Energy Intensive Manufacturing Industry in Sweden. Energy 31 1500-1508.

This paper treats barriers to energy efficiency in the non-energy intensive industry in Oskarshamn. The aim of the paper was to investigate the existence and importance of these barriers. The results highlight barriers inhibiting the diffusion of energy-efficient technology such as risk, cost of production disruptions, lack of time and other priorities, and lack of sub-metering. The study also found a number of drivers such as people with real ambition and the need for long-term strategy. This article was a result of a project within the National Energy Systems Programme.

Rohdin, P. Thollander, P. Solding, P. (2007) Barriers to and Drivers for

Energy Efficiency in the Swedish Foundry Industry. Energy Policy 35

672-677.

This paper evolved from the first study presented in paper one. In this paper the barriers to and drivers for energy efficiency within the energy-intensive foundry industry were studied. This project was made in collaboration with the Swedish Foundry Association. The results show that barriers within group-owned companies were more related to organizational problems and barriers within private foundries more related to information. The study also found that consultants and other actors working within the sector were considered more credible and thereby preferred before for instance governmentally funded energy audits. The most important drivers by far were found to be people with real ambition and long-term strategy.

(20)

Thollander, P. Danestig, M. Rohdin, P. (2007) Energy Policies for Increased Industrial Energy Efficiency - Evaluation of a Local Energy

Programme for Manufacturing SMEs, Energy Policy 35 5774-5783.

This paper is an evaluation of Project Highland, which was one of the largest local energy programs targeting small and medium-sized industry in Sweden. The study shows an adoption rate of approximately 40% for low- or no-cost measures within the studied industries. The paper also compares this energy efficiency program with another ongoing program targeting energy-intensive industry, and indications are that the approach used is effective. The paper also studies barriers to and drivers for energy efficiency. Lack of time and other priorities are shown to be the most important barriers followed by other priorities for capital investment. The main drivers found were long-term strategy followed by people with real ambition.

Rohdin, P. Moshfegh, B. (2007) Numerical Predictions of Indoor Climate in Large Industrial Premises - A Comparison between Different

k-ε Models Supported by Field Measurements. Building and

Environment 42 3872-3888.

This paper explores the benefits of using computational fluid dynamics as a method by which to predict indoor environment parameters in a complex industrial packaging facility. The paper also presents a comparison between three eddy-viscosity turbulence models: the standard k-ε model, the renormalized group (RNG) k-ε model and the realizable k-ε. The RNG model gave the results with least deviation compared with measured values. The paper also explored the potential energy savings of changing supply airflows and what consequences this would have on the indoor environment.

(21)

Rohdin, P. Moshfegh, B. (2006) Numerical Predictions of Indoor

Climate in a Light Alloy Casting Facility. Proc. of the 13th International

Heat Transfer Conference, IHTC-13, August 13-18, Sydney, Australia

This paper focuses on numerical simulation of indoor parameters and energy use using computational fluid dynamics and energy simulation. The main purpose is to evaluate the potential use of CFD when designing industrial ventilation systems in a foundry. The turbulence model used was the renormalized group k-ε model, and the results were compared with measurements. The paper shows a good agreement between simulated and measured values for both temperature and velocity in the 104 points measured. The paper also explores the impact of reducing supply airflow in the facility.

F. Karlsson, P. Rohdin, M-L. Persson, Measured and Predicted Energy Demand of a Low-Energy Building - Important Aspects When Using Building Energy Simulation, Services Engineering Research and Technology 28 (2007) 1-13.

This paper compares the use of three different energy simulation models. The simulated object was an empirically well-known low-energy building in Lindås in Sweden. The focus of the paper was on the choice of simulation software, the impact of the habits of the tenants, and the impact of the uncertainties in supply airflow, heat exchanger efficiency and internal loads. The paper shows small variation in predicted energy use for the different software programs. The most important parameters are shown to be the habits of the tenants when comparing the results with measured values. This underscores the importance of accurate design parameters during the design phase.

(22)

Rohdin, P. Moshfegh, B. (2007) CFD and BES as Decision Support When Implementing a Variable Air Volume System in a Foundry. In

Proc. of Clima 2007 WellBeing Indoor Congress, 10-14 June, Helsinki,

Finland.

This paper focuses on the combination of building energy simulation and computational fluid dynamics as a trustworthy and accurate method of obtaining detailed information when studying industrial ventilation and its control. The study includes predictions of the impact of implementing a Variable Air Volume (VAV) system in a light alloy casting facility. The paper shows that a VAV system is predicted to decrease energy and electricity use for industrial ventilation by about 30%.

The paper was awarded an honorary mention for best poster.

Rohdin, P. Moshfegh, B. (2007) Variable Air Volume-Flow Systems – A Possible Way to Reduce Energy Use in the Swedish Dairy Industry.

International Journal of Ventilation 5 381-392.

This paper explores the use of building energy simulations in predicting energy use and thermal comfort in the dairy industry. The focus is on industrial ventilation and an entire site is simultaneously simulated. The study includes a benchmark of two large dairy production sites built at three different stages and it is shown that the energy use per square meter increases with time, as does the electricity used for ventilation.

The use of a Variable Air Volume system with heat recovery is explored and shown to be an effective way to reduce both energy and electricity use. The main potential is in the reduction of heat used for space heating.

(23)

2 Swedish industrial energy use and some recent

studies of the efficiency potential

This chapter contains a short description of the Swedish energy system followed by an introduction to some large or recent studies of the energy efficiency potential in the Swedish industry.

he Swedish electricity generation system is based on a large amount of hydropower and nuclear power, and since the deregulation of the Swedish electricity market in 1996 the Swedish market has become more and more interconnected with the rest of Europe. This became even more so as the European Union accepted the EU Electricity Directive, which includes rules for the internal electricity market in the EU. The European system also has a different structure compared with the Swedish system. The European electricity system has a high degree of coal, natural gas and nuclear power.

Swedish industry accounts for 39% of the total energy use and 43% of the electricity use (SEA, 2005), and the energy use per energy carrier is presented in Figure 2.1, both for the total Swedish system and for the industrial sector.

T

(24)

Total Energy Use per Energy Carrier 131 47 133 7 17 66

Electricity District heating Oil products Natural gas Coal, coke Biofuels, peat, waste

Industrial Energy Use per Energy Carrier

57 5 21 5 17 51

Electricity District heating Oil products Natural gas Coal, coke Biofuels, peat

Figure 2.1. Total Swedish energy use and industrial energy use per energy carrier [TWh] (SEA, 2004).

The historical structure of energy use has changed even though the total industrial energy use has not increased very much in absolute terms despite significant increases in production. The main change is a drastic decrease in oil use since the 1970s, which has been compensated for by an increase in biofuels and electricity (SEA, 2004). Electricity, in contrast to oil, natural gas and coal, is not a primary fuel, and since primary fuels are used in the process of generating electricity, it will in all probability follow the general trend of the underlying fuels. The trend of increasing energy prices and other environmental policy instruments such as emissions trading systems and CO2 taxes raising the price of energy will change the conditions for industrial companies from what they are today. One way to reduce the risk of these increasing energy prices is to increase energy efficiency within the Swedish industrial sector. The question is then “How large is this potential for Swedish industry?”

There are a large number of publications on the potential for energy efficiency both international and domestic. However, the differences in market conditions, location, energy prices, legislation and environmental aspects make it hard to compare these studies. For this reason only more recent or large studies and programs are discussed, and for Swedish conditions only.

Perhaps the largest Swedish energy program, called “Uppdrag 2000” (Commission 2000), was reported in four parts, with one part treating the industrial sector. The study targets Swedish industries with more than four employees, thus including the food industry, wood industry, graphical industry, textile industry, manufacturing industry, and parts of the chemical and

(25)

quarrying industry. The general conclusions of this large program include issues such as: (1) Electricity conservation can be achieved through more efficient utilization of existing technology and modifications of equipment to the actual requirements; (2) Oversizing is common, especially for electric motors; (3) The importance of a whole system perspective together with careful operation and maintenance instructions are important in reducing energy use; (4) The greatest potential to reduce electricity use is found in connection with constructing new buildings, repair and renovation; (5) Measures that simultaneously result in other positive spin-off effects are more likely to be implemented, such as better working conditions, improved indoor climate and improved operational reliability. The final report from Uppdrag 2000 was published in November 1991.

In January 2005, a program called Program for Energy Efficiency (PFE) was initiated by the Swedish Energy Agency. The PFE is a program targeting electricity use in energy-intensive industry, and presently more than a hundred companies are active in the program. The companies participating in the program receive a tax reduction, provided that they work strategically with the energy issue and implement efficiency measures within their companies (SEA, 2007). During the first two years 98 companies participated, representing an electricity use of 29 TWh/year, and about 900 efficiency measures with an annual potential of 1 TWh have been generated. The program includes energy audits in the initial phase and also the implementation of an energy management system. For these energy-intensive industries, 48% of the measures were related to the production process, leaving 52% of the measures to support processes. Among the support processes, compressed air, pumps and fans are the processes with most measures suggested (SEA, 2007). It is important to note that this program's main goal is to target electricity. Results from this program are also presented in Ottosson and Petersson (2007).

A large energy auditing program in Sweden called Project Highland, covering about 340 companies of which 139 are industrial companies is currently being undertaken. This program targets small and medium-sized industrial companies in the Highland region in Sweden. A large portion of the potential cited is related to heating ventilation and air-conditioning of the industrial premises, as nearly all of the possible heat conservation is related to space heating and additional electricity used by cooling compressors and fans. An evaluation has

(26)

been made of the first part of the program and is found in Thollander et al. (2007).

In several other recent Swedish studies such as Trygg and Karlsson, 2005; Nord-Ågren, 2002; Dag, 2000; Thollander et al., 2005; Karlsson et al., 2004; Henning, 2005, the technical potential for reducing the energy use in Swedish industrial companies have been studied. In Trygg and Karlsson (2005), a 40% reduction in energy use and 48% reduction in electricity use are possible for the 11 industries studied. In Thollander et al. (2005), the potential reduction was assessed at 33% for energy and 23% for electricity at a foundry. In Karlsson et

al. (2004), two large dairy processing plants were studied, and the potential for

reducing energy use was 8% of these measures; 32% was related to HVAC, which indicated a large potential for reducing energy and electricity use in HVAC. In a summary of energy auditing projects in Sweden by Henning (2005) the potential for 10 industrial sectors based on 40 energy audits is presented. The average potential for these sectors is about 40% for energy and 45% for electricity. The main potential for reducing the energy use is found in support processes such as space heating (and cooling), ventilation, lighting and compressed air. In another industrial study of a large car manufacturing plant by Trygg et al. (2006) focusing on the efficiency potential for electricity, the total amount of energy used by support processes is 59.1 GWh (of which 34.5 GWh is electricity) and of that space heating and ventilation represents 55%. The potential for reducing the electricity used by fans by more efficient control including using a variable air volume systems is found to be about 35%. The impact on heat demand was not assessed.

The variation in energy efficiency potential in the studies presented above ranges from about 10% to 50% for the different cases. There are several factors making the audits hard to compare, such as different rates of return on invested capital, different structure of energy use at the sites, different goals as some of the studies or programs only focus on electricity while others have a broader approach and target energy, some have environmental focus and target CO2 -effective measures while others target cost--effective measures. However, whatever the focus, the studies all find substantial potential for implementing effective measures and the studies result in an increase in implementation at the firms studied. The studies have another aspect in common: they indicate that the diffusion of investment in efficient HVAC installations seems to be slow

(27)

and thus a large potential for cost-efficient investments in this area is currently being overlooked. Mechanisms inhibiting this potential from being implemented are discussed in Chapter 3 where some industrial management issues related to energy management are briefly covered.

(28)

(29)

3 Industrial Management Issues

Industrial management issues and issues related to decision making are important when studying reasons to why energy efficiency measures are or are not undertaken. In this chapter the used theory of barriers to energy efficiency as well as a short description of rational decision making is included.

3.1 Rational decision making

he management of an industrial company requires decisions related to e.g. management of production, maintenance, facilities, energy and labor issues. Within each of these categories there are multiple dimensions to deal with, such as risk evaluation, what information on which to base the decision, how to finance different measures, education and the hiring of personnel (Sandberg, 2004). Management and its strategy are strongly related to the investment decision the organization has to make in order to be profitable, which is the overall goal of a company. This company strategy includes being alert to changing market conditions and other competitors and suppliers as well as issues like environmental concerns. The ability to make the “correct” decision, in relation to the overall goal, is often stated as what separates a successful business from an unsuccessful one. When making these decisions, if not made by rule of thumb, resources have to be allocated to the process of evaluating competing alternative investments. These resources, in the form of personnel, information, audits, simulation tools and consultants, are needed to build a sound foundation for decision making (Sandberg, 2004).

An investment decision usually involves choosing between a number of alternative solutions, each of which has consequences for the future. In many

T

(30)

corporate and technical models, the theory of rational decision making is used. In this theory the rational decision maker has the ability to choose the optimal solution to the problem, based on a decision process where:

1. The problem is explicitly defined and the goal is unambiguously stated 2. All possible alternatives are gathered and examined

3. All the consequences of all these alternatives are established 4. The consequences are compared with the overall goal

5. The alternative with the highest degree of fulfillment in relation to the goal is chosen

When using this procedure of rationality the decision maker is confronted with several problems. All the alternatives have to be known, it must be possible to know the consequences of all these alternatives, there must be an unambiguous goal and the different alternatives must be comparable in all aspects. The decision maker is faced with an impossible task, which is why investment decisions in all probability are not made strictly in this fashion (Andersson, 1997). This is elegantly put by James G. March (1998) as: “Rational choices involve two kinds of guesses: guesses about future consequences of current actions and guesses about future preferences for those consequences.”

Theories of choice often assume that future preferences are stable and known with sufficient precision to make decisions unambiguous (March, 1998), which in some cases is a rather large simplification. Another problem with models of individual choice, such as the “economic man”, is that the model doesn’t predict the behavior of any individual. This problem can however be avoided to a large degree with the use of aggregation of a large number of persons or organizations (March, 1998).

The concept of rationality is not connected only to the economic field but also to other fields such as psychology, sociology and political science (March, 1988). The concept of bounded rationality is introduced by H. Simon where he argues that decisions are made with bounded rationality, where decision makers are only able to recognize a limited number of alternatives, and are aware of only a number of the consequences. He argues that human abilities are fallible and limited, information is never perfect, and that among other things, time

(31)

and capital are limited. As a result it is thus impossible to maximize benefits in decision making; instead a satisfying decision is to be sought (Torsi et al., 1994). In addition to the economic considerations the final decision should include additional aspects such as work environment issues, labor security issues and liquidity aspects, as these are hard if not impossible to include in a strict economic model (Andersson, 1997).

3.2 Barriers to energy efficiency

There are numerous studies stating the existence of a gap between potentially cost-effective energy efficiency measures from a technical perspective and the measures actually implemented (DeCanio, 1993; DeCanio, 1998; Hirst and Brown, 1990; Howart and Andersson, 1993; Jaffe and Stavins 1994a-b; Ramirez

et al., 2005; Sanstad and Howarth, 1994, Sorrell et al., 2004 and Weber, 1997).

This gap has been referred to as the “energy efficiency gap”, and is argued to exist due to the presence of a number of barriers to energy efficiency preventing cost-efficient energy efficiency measures to be implemented. The energy barrier framework is mostly derived from mainstream economic theory and a deviation from this theory is thus explained in terms of a barrier.

One purpose of introducing a concept of barriers is to assign explanatory variables to why the companies/decision makers act as they do and thus give policy makers means to efficiently target their behavior. These theories are thus argued to result in better understanding of the internal function of the firm and consequently allow better modeling. However, it is important to note that the barriers are not unambiguous and that an empirical find may be related to more than one barrier.

According to Sorrell et al. (2000), a barrier is defined as:

“A postulated mechanism that inhibits investments in technologies that are both energy efficient and economically efficient.”

The barrier theory, used in Papers I, II and III, uses the distinctions of Sorrell et

al. (2000) where the barriers are divided into two economic, one organizational,

(32)

Furthermore, it’s worth noting that the theory has a “company perspective” and the barriers are used to explain the corporate view on energy efficiency measures, not the policy makers’ view of the company. Outside the company there may exist other explanatory variables for not adopting a measure.

The theoretical barriers in the framework used are hidden cost, limited access to capital, risk, heterogeneity, adverse selection, principal agent relationship, split incentives, imperfect information, form of information, credibility and trust, value, inertia, bounded rationality, power and culture. A more comprehensive description of the theoretical barriers used in the appended papers is found in below.

Hidden costs

Hidden costs are defined as costs associated with information seeking, meeting with sellers, writing contracts, etc. These costs are often used to explain large parts of the energy efficiency gap (DeCanio, 1998). The argument is that when included, the otherwise cost-efficient measure is no longer cost efficient since the hidden costs stand for a significant part of the actual profit of an implementation. Hidden costs are a frequently used argument against the existence of an energy efficiency gap, by arguing that engineering-economic studies often fail to see the full cost of an energy efficiency measure (Sorrell et

al., 2000).

Limited access to capital

Energy-efficient equipment often has a higher initial cost than less efficient equipment. This may have multiple reasons and some are discussed in Almeida (1998), which we will refer to later when discussing results from the studies. Limited access to capital is a barrier when the additional capital needed to make an energy-efficient investment is hard to obtain, even though the investment is cost efficient in terms of the company’s investment criteria (Hirst and Brown, 1990).

Heterogeneity

A population of firms is most likely heterogeneous with respect to their energy use, which is why a technology that appears cost efficient on average will be cost inefficient for some portion of the population (Jaffe and Stavins, 1994).

(33)

This barrier is arguably larger for industrial companies where the processes and plants are to a large degree site specific. This is one strong argument for providing site-specific information to increase adoption of energy-efficient equipment, such as on-site energy audits and investigations instead of general information.

Risk

In Sorrell et al. (2000) risk is divided into three categories: external risks such as overall economic trends and expected reductions in fuel prices; business risk, including sector-specific economic trends; and technical risks, such as reliability and technical performance of a specific technology. Risk is also often used to explain shorter pay-back periods or high discount rates when investing in energy-efficient equipment compared to other production related investments. In Hirst and Brown (1990), it is explained that even though managers know the cost of an energy-efficiency investment, uncertainty about the long-term savings in operating costs means the investment is a risk as future energy prices are not known. The uncertainties in future energy prices and availability are argued in Stern and Aronson (1984) to be an important aspect increasing the risk of investments in energy efficient equipment. This and the irreversibility of the investment, also discussed in March (1988), in energy efficiency technology are argued to be an explanation for applying relatively high discount rates for investments whose return is uncertain (Jaffe and Stavins, 1994). In the concept of risk, the risk of trying new equipment which may lead to an increased amount of production disruptions is also included and has been cited in Hirst and Brown (1990) to be very important to decision makers.

Adverse selection

When the two parties in a transaction have had access to different levels of information, the problem of asymmetric information arises. The problem is extremely common in the real world, and Sanstad and Howarth (1994) argue that this is more common than not. Manufacturers and retailers of energy-efficient equipment are often much better informed about the performance than the buyer which represents an adverse selection.

Principal-agent relationship

The principal-agent relationship problem is another form of asymmetric information which is due to lack of trust between parties in different levels

(34)

within an organization. The manager, who may not be as well informed about energy efficiency investments, may demand extra short pay-back rates on energy efficiency investments due to distrust in the ability of the organization/department to make such an investment, inhibiting cost-efficient energy efficiency investments (DeCanio 1993; Jaffe and Stavins, 1994).

Split incentives

In Jaffe and Stavins (1994) the barrier of split incentives is argued to be important when the party receiving or discovering an energy efficiency investment is not the party that pays the energy bill. Then a cost-efficient measure may not be sufficient for adoption. The adoption will only occur if the adopter can recover the investment from the party that has the incentive to save energy (Jaffe and Stavins, 1994). In Hirst and Brown (1990), it is argued that this is a problem for energy-efficient technologies, which often have higher initial costs but lower life-cycle costs than conventional technologies. A common type of split incentive is the landlord-tenant relationship. Lack of sub-metering within multidivisional organizations may also be classified as a split incentive.

Imperfect information

That consumers are poorly informed about market conditions, technology characteristics and their own energy use is often cited. In Sanstad and Howarth (1994) the authors argue that lack of adequate information about potential energy-efficient technologies inhibits investments. Imperfect information may be divided into lack of information, where information of energy performance may not be available to the buyer; cost of information, when there are costs associated with searching and acquiring information about energy performance; and accuracy of information, where the information provider may not always be transparent or truthful about the energy performance. In Sorrell et al. (2000) it is argued that imperfect information is likely to be more common for adoption of products purchased infrequently. Problems related to imperfect information are often countered with information campaigns.

Form of information

There is a relation between the form of information and the degree of adoption of suggested measures. When information is provided in an inefficient form it is thus argued to constitute a barrier. In Stern and Aronson (1984) as well as in

(35)

other sources it is shown that people are more likely to remember information if it is specific and presented in a vivid and personalized manner coming from a person that is similar to the receiver. This has implications for how policy instruments such as an energy audit or a complex investment assessment should be performed, indicating increased adoption for site-specific, vivid information such as a presentation of detailed results from measurements or simulations.

Credibility and trust

A prerequisite for efficient adoption is the adopters’ perceived trust in the information provider. In Stern and Aronson (1984), it is argued that efficiently spread information and trust in the information provider is essential.

Values

Our values influence our behavior and in studies of households it is cited that value a have strong impact on cost-free conservation measures, but a weaker impact on low-cost conservation measures. The behavior of surrounding people also has great impact on the behavior of a person, as friends and colleagues implementing energy efficiency equipment or with conservation behavior act as good examples (Stern and Aronson, 1984).

Inertia

This concept states that individuals and organizations are, to some degree, “beings” with habits and tend to establish routines and stick to them. This is argued in Stern and Aronson (1984) to be a way for decision makers to reduce perceived uncertainty and change in their surrounding by avoiding or ignoring problems. Stern and Aronson (1984) also show that a person who has made an important decision seeks to justify and rationalize this by convincing themselves and those around them that the decision was the correct one. Inertia becomes a barrier when energy users and decision makers fail to take economically justifiable action to save energy as a result of not reevaluating routines and procedures (Stern and Aronson, 1984).

Bounded rationality

Bounded rationality is connected to problems related with multiple objectives for individuals as well as organizations and imperfect information etc. when making decisions. In Sanstad and Howarth (1994) this concept is applied to

(36)

energy-related decisions and it is found that this is an important explanatory variable, as the assumption of rational decision makers would require individuals and firms to solve extremely complex optimization problems in order to obtain the optimal energy service. Furthermore, a company does not consist of one person with one view, but multiple individuals with their own views and the interests of one individual or department may conflict with other departments’ or individuals’ interests. Also, organizations or individuals do not act on the basis of complete information but rather make decisions to some degree by rule of thumb (Stern and Aronson, 1984).

Power

In many organizations energy management has low status compared to for instance the production process. This is argued to be a barrier to energy efficiency when this low status leads to constraints when striving to implement energy efficiency measures Sorrell et al. (2000).

Culture

In Sorrell et al. (2000) culture is explained as the sum of each individual’s values where the values of executives or other workers have influence within the organization. Culture is not by definition a barrier, and it is closely connected to value. But even though culture is not a true barrier it has been shown to have significant impact on the adoption of energy efficiency measures (Sorrell et al., 2000).

(37)

4 Indoor environment and energy use in industrial

buildings

In this chapter Indoor Air Quality, Thermal Comfort, Product environment issues and Energy use for HVAC are described, and used measures of performance are also included.

he indoor environment, product environment and energy use in industrial facilities are closely linked. The term “indoor environment” includes both the thermal environment and the indoor air quality (IAQ). For industrial facilities it is also common to treat the product environment separately. At the end of this chapter energy issues related to HVAC will be discussed.

4.1 Indoor air quality (IAQ)

The quality of air in indoor areas occupied by humans is often referred to as the Indoor Air Quality (IAQ). Fanger (2006) defined IAQ as “The extent to which

human requirements are met” (in terms of air quality). This expresses whether the

air is perceived as fresh and pleasant, that it doesn’t have a negative effect on human health, or that the air doesn’t negatively affect the ability to work or think. Fanger (2006) argues that when examining IAQ for non-industrial buildings, controlling the quality of the air by assigning target values for all known chemicals reducing the IAQ is not possible, as there are typically thousands of compounds in the air in very low concentrations which affect the IAQ. However, when controlling IAQ for industrial buildings, the concept of target values for emissions is an often-used design tool, as often only a few compounds are substantially more important and can thereby be used as design criteria (Goodfellow and Tähti, 2001). Target values for numerous common compounds are found in ASF 2005:17 – Occupational exposure limit values

T

(38)

and measures against air contaminants (AFS, 2005) issued by the Swedish Work Environment Authority. The hygienic target values are usually divided in three categories, a Level limit value, a Ceiling limit value and a Short-term value. The Level limit value refers to the average exposure during a work day. The Ceiling limit value refers to an average value of exposure during a specific amount of time, often either five minutes or 15 minutes. The short-term value is a recommended value which is based on a reference period of 15 minutes. The difference between the ceiling value and the term value is that the short-term value is more of a guideline than the ceiling limit value (Hedlund, 2005). In international literature such as ACGIH (2004) the health hazard of an airborne compound is categorized by the Threshold Limit value (TLV) which, according to (ACGHI, 2004), “is defined as that airborne concentration of a substance which it is believed that nearly all workers may be exposed to day after day without developing adverse health effects.” If multiple compounds with similar effects are simultaneously present, the effect of these should be evaluated together when designing the ventilation. In AFS (2000) it is stated that the concentration of contaminants in the air should be as low as possible when considering the specific production process. This means that even if the concentration is below the level limit it is to be decreased if that can be done using reasonable economic means (Goodfellow and Tähti, 2001). There are, however, some important limitations to this concept, as all the contaminants need to be known and new toxicological aspects when mixing different contaminants must also be known. This is not always treated specifically by a target level assessment in its common form.

4.1.1 Measures of performance

The performance in terms of IAQ for polluted production facilities is recommended to use target level assessments, and they are to be evaluated in terms of measured concentration, or when appropriate or necessary with calculations of the emission concentration. In non-industrial facilities the concept of using human sensory response is suggested to determine the perceived IAQ, measured in decipol or percentage dissatisfied (PD) (Fanger, 2006). This concept is argued to also be a workable measure of performance for non-production contaminated industrial facilities which are similar to

(39)

facilities such as residential buildings or offices. The PD index for non-production contaminated industrial premises may then be defined as:

) 25 . 0 83 . 1 ( exp 395 v PD= ⋅ − &

(4.1)

where

v& is the outdoor airflow rate in ls-1olf-1, see Awbi (2003). 4.1.2 Requirements

The requirement for Swedish companies is to fulfill the regulations issued by the Swedish Work Environment Authority, by providing an indoor air quality not exceeding any of the emissions cited in the target levels in AFS (2005). The requirements should also fulfill other forms of legislation and GMPs (Good Manufacturing Procedures) of specific sectors. As an example, for the dairy studied in Papers IV and VIII, the requirements stated in the Food and Drug Administration's regulations should also be taken into consideration (NFA, 1996) and, if found applicable, recommendations such as those issued by Brown (1996). Extended site-specific requirements are sometimes used to improve indoor air quality to promote a comfortable and healthy working environment.

4.2 Thermal comfort

Humans want and seek thermal comfort wherever they are, whether at work or at home. The term thermal comfort may be defined as: “that condition of mind that

expresses satisfaction with the thermal environment” (ASHRAE, 2003). Clothing,

activities, posture, location and shelter are chosen, altered and adjusted to minimize discomfort both consciously and unconsciously. Discomfort is related to human health, number of mistakes, productivity and industrial accidents, as discomfort is a form of physiological strain. This strain can be in the form of sweating, muscle tension, stiffness, shivering and loss of dexterity (Goodfellow and Tähti, 2001). Thus, thermal climate is an important factor and it is desirable to minimize thermal discomfort. In general it is often cited that a facility that provides a satisfactory environment will be financially more successful as well as more desirable for the people working there, as it contributes to a safer and more comfortable environment with less loss of productivity and better comfort (Goodfellow and Tähti, 2001).

(40)

When making statistical correlations of comfort in the thermal environment one may distinguish some reproducible primary factors and some secondary factors of less importance. The lesser factors are factors such as age, local climate, physical fitness, food and illness. These secondary factors have a lesser effect on the perceived thermal climate and are not discussed further in this thesis. The primary factors directly affect the heat transfer of the person, where the objective of a person’s thermoregulation is to maintain the body core temperature vital for the internal organs to function properly (Goodfellow and Tähti, 2001).

One way to quantify comfort and discomfort is through empirical environmental indices. These studies are based on a model of the heat exchange between the body and its surroundings, which is made by subjecting a large population to different known activities and clothing to a range of environmental conditions, and recording the response to these conditions. Probably the most used model is the Predicted Percentage Dissatisfied (PPD) index created by Fanger and his co-workers in Denmark (Fanger, 1972). This work as well as other more zone-specific indices is included in the ISO 7730 standard of “Ergonomics of the thermal environment” (ISO, 7730). To compare the comfort sensation of different thermal climates in this thesis, the concept of empirical environmental indices will be used. In Papers IV and V, the concept of Predicted dissatisfied due to draught was used as this was considered to be a problem in the studied facility. In Paper VIII several of the indices presented below were used to predict comfort. The indices that have been used as measures of performance related to comfort are summarized below.

4.2.1 Measures of performance

To quantify the consequences in terms of comfort in this thesis three major indices are used, the PDDraught (Predicted Dissatisfied due to draught), PMV (Predicted Mean Vote), and its related PPD index.

To relate the physical parameters such as temperature, velocity, and turbulence intensity to the predicted comfort of the personnel, the PD Draught index has been used. The PD Draught index is defined as:

(41)

) 3696 . 0 143 . 3 ( ) 05 . 0 )( 34 ( 0.6223 Draught toz Uoz Uoz TIoz PD = − − + ⋅ (4.2)

where PDDraught represents the percentage of dissatisfied people due to draught in the head region. This has been used in several of the appended papers and has been implemented in the CFD code (FLUENT) using a custom field functions. For PDDraught <0, which occurs when Voz< 0.05 m/s the PDDraught has been set to zero to avoid undefined nodes.

In Fanger (1970) human thermal comfort has been described by the predicted mean vote (PMV) index. The PMV index represents the mean votes of a large group of individuals describing their thermal sensation using a seven-point thermal scale that ranges from hot to cold, with neutral at zero (ISO, 1994; ISO, 2005). The equation for the PMV index is defined by:

(

)

(

)

[

(

)

]

(

)

[

]

(

)

(

)

(

) (

)

[

]

cl c

(

cl a

)

]

4 r 4 cl cl 8 a a 5 a 3 036 . 0 273 273 10 96 . 3 34 4 001 . 0 5867 10 7 . 1 15 . 58 42 . 0 99 . 6 5733 10 05 . 3 028 . 0 303 . 0 PMV t t h f t t f t M p M W M p W M W M e M − − + − + × ⋅ − − ⋅ − − ⋅ − − − × − − − − × ⋅ − × + = − − − − (4.3) where,

(

)

{

[

(

) (

)

]

(

)

}

(

)

(

)

(

)

⎪⎩ ⎪ ⎨ ⎧ ⋅ > + ⋅ ≤ + = ⎪⎩ ⎪ ⎨ ⎧ < − > − − = − + + − + ⋅ − − − = − W C I I W C I I f t t t t t t h t t h f t t f I W M t / m 078 . 0 for 645 . 0 05 . 1 / m 078 . 0 for 290 . 1 00 . 1 1 . 12 38 . 2 for 1 . 12 1 . 12 38 . 2 for 38 . 2 273 273 10 96 . 3 028 . 0 7 . 35 o 2 cl cl o 2 cl cl cl ar 25 . 0 a cl ar ar 25 . 0 a cl 25 . 0 a cl c a cl c cl 4 r 4 cl cl 8 cl cl ν ν ν

It is also possible to transform the PMV index into a statistical expression for the Predicted Percentage of Dissatisfied (PPD) people. The expression for the PPD index is based on empirical correlations and the equation reflects the fact that in large groups of people there will be at least about 5% dissatisfied no matter how the overall thermal climate is arranged. The expression for the PPD index contains only the PMV of the group and is given by:

( 4 2) PMV 2179 . 0 PMV 03353 . 0 95 100 PPD= − ⋅e− × + × _(4.4)

(42)

4.2.2 Requirements

In the standard for the PMV and PPD index (ISO 7730, 1994; ISO 7730, 2005) it is suggested that the requirements for the physical variables: temperature, mean radiant temperature, relative air velocity, and partial water vapor pressure should be solved for PMV=0. This is because this is when most people are predicted to be neutral in terms of their climate sensation, neither to hot nor to cold. The use of this concept may however result in large installations and high energy use. However, in the newer standard (ISO 7730, 2005), this concept is less in focus.

The use of the controlled indices instead of more adaptive models for free running buildings in this thesis is due to the occupants’ inability to control their climate. For the industrial cases studied in this thesis the occupants have been entitled to wear certain types of clothing, special clean clothing and hairnets in the dairy case and boiler suits in the foundry case. Furthermore, the occupants don’t have any way to control the indoor temperature or airflows in the premises.

4.3 Product environment

The product environment is defined as the surrounding environment that the product requires in order to be of sufficient quality, and is specific for each process and may need to be investigated for each individual case. However, many manufacturing and storage processes don’t have any extended product requirement. In Papers IV and VIII, a dairy processing plant has been studied and for these types of plants there are several extended requirements in order to produce products of high quality. For instance, it is recommended that the airflow between different product facilities should be in the direction from clean to less clean areas. The air filters should also be easy to reach and procedures for changing them are required. Guidelines on air quality for the food industry may be found in publications such as Brown (1996). Here issues such as internal condensation are discussed and it is recommended that ventilation and process equipment be designed to avoid condensation since this leads to extended microbiological growth. For the foundries such as those discussed in Papers V and VII the issue of condensation on the metal to be

(43)

melted may be of interest. In addition it is also recommended to have a under pressure in this type of facility, to avoid unnecessary transport of contaminants into the ambient facilities.

4.4 Energy use in HVAC

There are different ways in which to categorize and present energy use and energy efficiency indicators. In Jaegemar (1996) energy efficiency is defined as:

“Energy efficiency is a measure of the balance between the energy gained and the sacrifice necessary to bring about this gain”. An energy-efficient system is thus a system that

provides its function with the lowest possible energy use at reasonable cost. This definition is useful when discussing indoor environment aspects as when Abel and Ekberg (2002) stress the importance of two aspects when implementing efficiency measures:

1. The implementation of an energy efficiency measure should not have a negative effect on the building and its function.

2. The primary use of resources when implementing an energy efficiency measure should be related to the total energy reduction due to the measure.

These two requirements are to be fulfilled in order for the measure to be categorized as an efficiency measure. This sorts out measures where the function of the building, such as IAQ issued, is reduced or when the measure results in a use of resources not in proportion to the reduction of energy use (Abel & Ekberg, 2002). Another aspect in defining an energy efficiency measure could be to establish an energy pay-back criteria, where the energy pay-back index represents the potential saving per year divided by the embodied energy, and thus results in a relation of how long it will take for the measure to save as much energy as it has required in production.

In Jagemar (1996) a thesis on designing energy-efficient HVAC installations targeting the explicit energy performance indices of commercial buildings is presented for individual components in the HVAC system. Here the energy performance ratios are defined in terms of kWh/(m2_{and year), kWh/(m}3_and year), kWh/(m3_{/s) and kW/(m}3_{/s) at different levels in a building.}

Energy efficiency and ventilation in Swedish industries